The adoption of large language models (LLMs) in healthcare demands a careful analysis of their potential to spread false medical knowledge. Because LLMs ingest massive volumes of data from the open Internet during training, they are potentially exposed to unverified medical knowledge that may include deliberately planted misinformation. Here, we perform a threat assessment that simulates a data-poisoning attack against The Pile, a popular dataset used for LLM development.
View Article and Find Full Text PDFOper Neurosurg (Hagerstown)
December 2024
Background And Objectives: Superficial temporal artery to middle cerebral artery (STA-MCA) bypass is the workhorse for flow augmentation surgery. Although either interrupted or running sutures can be used to complete the anastomosis with high intraoperative patency rates, no previous study in the cranial bypass literature has compared long-term patency and maturity of end-to-side STA-MCA anastomoses. We compared STA-MCA anastomoses performed with running vs interrupted sutures by evaluating bypass flow and anastomotic maturation on follow-up vascular imaging.
View Article and Find Full Text PDFOper Neurosurg (Hagerstown)
September 2024
Background And Objectives: Iatrogenic cerebrovascular injury can cause intracranial hemorrhage and pseudoaneurysm formation, putting patients at high risk for postoperative bleeding. No consensus for management exists. This study describes endovascular treatment of these acute injuries with flow diverter stents.
View Article and Find Full Text PDFObjective: The objective of this study was to report the results of a bibliometric analysis on the modern corpus of literature pertaining to endoscopic third ventriculostomy (ETV). Prior bibliometrics studies on ETV have focused on highly cited articles, but an advanced bibliometric analysis has not yet been conducted.
Methods: The authors queried the Web of Science (WoS) for (ALL = (endoscopic third ventriculostomy)) OR (ALL = (ETV) AND ALL = (neurosurgery)).